The objective of this study was to compare the risk of estimated glomerular filtration rate (eGFR) decline between atrial fibrillation (AF) patients with direct oral anticoagulants (DOACs) and warfarin.
We studied patients with nonvalvular AF from a prospective multicenter national AF registry in Thailand. Patients with missing eGFR data or eGFR less than 30 mL/min/1.73 m2 were excluded. Follow-up data including eGFR were collected every 6 months until 3 years. eGFR decline was assessed by eGFR slope. We compared eGFR slope between patients who received DOACs and warfarin at baseline. In the warfarin group, we assessed the impact of good anticoagulation control by time in the therapeutic range (TTR).
A total of 1708 patients were studied (mean age 68.1 years; 42.6% female). Patients with DOACs had a significantly slower rate of eGFR decline compared to warfarin. The eGFR slope was 2.32 mL/min/1.73 m2 per year in the warfarin group (95% CI: 3.09 to 1.55), and 1.31 mL/min/1.73 m2 per year in the DOAC group (95% CI: 1.97 to 0.64). The effect of OAC type on the eGFR slope remained significant even after the adjustment of baseline variables including baseline eGFR. There was no difference in GFR decline as reflected by eGFR slope when comparing warfarin patients with TTR <65% and ≥65%.
In this prospective cohort of Asian patients with AF, DOACs were associated with a slower rate of eGFR decline when compared with warfarin. In the latter group, this was irrespective of the quality of anticoagulation control.